While it is clear that cytoskeletal elements (actin and microtubules) play roles in the dynamics and integrity of intracellular organelles, the mechanisms by which these actions occur are uncertain. In the last grant period, my laboratory revealed exciting cellular functions for the mammalian formin protein INF2. Importantly, INF2 exists as two C-terminal splice variants that essentially act as distinct proteins. One splice variant, INF2-CAAX, is prenylated and tightly bound to endoplasmic reticulum (ER). We showed that INF2-CAAX functions in mitochondrial fission. INF2-CAAX polymerizes actin filaments at the ER/mitochondrial interface, and INF2-mediated actin polymerization stimulates recruitment of the dynamin-like GTPase Drp1 to mitochondria, leading to fission. A second INF2 splice variant, INF2- nonCAAX, is not prenylated and is found in an actin-dependent meshwork-like pattern in the cytoplasm. INF2-nonCAAX plays a role in generation of dynamic actin patches in the Golgi region, and our results suggest that these patches act in fission of Golgi and other organelles. Overall, our results suggest that INF2 is a novel actin-based membrane fission factor. In this grant period, we study how INF2's effects on actin translate into mitochondrial fission. Aim 1 addresses the biochemical mechanism by which INF2 influences actin dynamics. INF2 is an unusual formin biochemically. While most formins accelerate actin polymerization, INF2 accelerates both polymerization and depolymerization. To depolymerize filaments, INF2 possesses a potent filament severing activity. In this aim, we elucidate INF2's severing and depolymerization mechanism using TIRF microscopy, Atomic Force Microscopy and biochemical techniques. We also use mutagenesis to determine important sequences for severing/depolymerization, and will use these mutants for experiments in Aims 2 and 3. Aim 2 addresses the mechanism by which INF2 deforms membranes. Based on our preliminary results showing that mitochondrial fission requires myosin activity, we hypothesize that INF2-generated filaments are tracks for myosin-mediated contraction. In this aim, we use fixed-cell and live-cell microscopy to test this model. Aim 3 addresses INF2 regulation. While other formins are regulated by auto-inhibition, our preliminary data suggest that an additional molecule is required for INF2 inhibition. Furthermore, we have evidence that Drp1 might be an inhibitory factor for INF2-CAAX. We use purified proteins and a novel cell-free assay system to elucidate Drp1's effect on INF2 and on actin, as well as to identify inhibitors for INF2-nonCAAX.